Face Recognition in Low Resolution Images. Trey Amador Scott Matsumura Matt Yiyang Yan

Transcription

1 Face Recognition in Low Resolution Images Trey Amador Scott Matsumura Matt Yiyang Yan

2 Introduction

3 Purpose: low resolution facial recognition Extract image/video from source Identify the person in real time given a traineddatabase taken from

4 Face Recognition Libraries histogram of oriented gradients (HOG) dlib Support Vector Machines (SVM)

5 Process Neural Enhance library increase the resolution of low pixel density Theano (neural network) Lasagne (train) upsampled image dlib Histogram of oriented gradients (HOG) SVM feature descriptor for detecting faces

6 Database IMDb Internet Movie Database is an online database of information related to films, television programs and video games low and high resolution versions of the same image highresolution 'base' image to train the Support Vector Machine (SVM)

7 Support Vector Machine for Face Recognition Arnold Schwarzenegger

8 SVM Identify the Rock Image of Images similar to

9 SVM The Rock not The Rock Images similar to Image of

10 SVM Separate data Images similar to Image of

11 SVM Which line? Images similar to Image of

12 SVM Thickest line Images similar to Image of

13 SVM Separate data? Images similar to Image of

14 SVM Nonlinear separation Images similar to Image of

15 Generative Adversarial Network for Upsampling Images

16 GAN Back with The Rock Image of Images similar to

17 GAN Generate this image? Image of Images similar to

18 Generative Network produce an image Discriminative Network real or fake vs

19 How to train your Generative Adversarial Network

20 GAN Train discriminative network real Discriminative Network fake

21 GAN Train both networks random noise Generative Network Discriminative Network Fake negative gradient positive gradient backpropagation

22 GAN Eventually? random noise Generative Network Discriminative Network Real backpropagation

23 GAN Upsampled Generative Network Discriminative Network Real

24 Code can be found at:

25 super resolution video samples

26 face recognition in enhancedresolution video

27 super resolution image enhancement boring Bruce Springsteen 100 x 100 enhanced Bruce Springsteen 200 x 200 actual Bruce Springsteen high res

28 super resolution face recognition unrecognized Bruce Springsteen 100 x 100 that s Bruce Springsteen! 200 x 200

29 experimental paradigm true face high res low res enhanced res false face high res low res enhanced res

30

31

32

33

34 future directions find robust metric with which to filter data test efficacy of various algorithms generate larger dataset

35 References [1] W. Zhao, et al. Face Recognition: A Literature Survey. ACM Computing Surveys, vol. 35, pp , Dec [2] S.C. Park, M.K. Park, and M.G. Kang. SuperResolution Image Reconstruction: A Technical Overview. IEEE Signal Processing Magazine. May [3] D. Glasner, S. Bagon, and M. Irani. SuperResolution from a Single Image, in IEEE 12th ICCV, 2009, pp [4] W.W. Zou and P.C. Yuen. Very Low Resolution Face Recognition Problem. IEEE Transactions on Image Processing, vol. 21, pp , July [5] A. Geitgey, "Face Recognition," GitHub repository, [Online]. Available: [Accessed ]. [6] N. Dalal and B. Triggs. Histogram of Oriented Gradients for Human Detection in CVPR, 2005, pp. 18. [7] P. Felzenszwalb, et al. Object Detection with Discriminantly Trained Part Based Models. IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 32, pp , Sept [8] C. Cortes and V. Vladimir, "SupportVector Networks," Machine Learning, vol. 20, no. 3, pp , [9] A. J. Champandard, "Neural Enhance," GitHub repository, [Online]. Available: [Accessed ]. [10] D. G. Lowe, "Object Recognition from Local ScaleInvariant Features," Computer Vision, vol. 2, pp , [11] K. Simonyan, M. O. Parkhi, A. Vedaldi and A. Zisserman, "Fisher Vector Faces in the Wild," British Machine Vision Conference, vol. 2, no. 3, p. 4, Sept [12] P. Fischer, A. Dosovitskiy and T. Brox, "Descriptor Matching with Convolutional Neural Networks: a Comparison to SIFT," arxiv, p. 10, 22 May [13] M. O. Parkhi, A. Vedaldi and A. Zisserman, "Deep Face Recognition," British Machine Vision Conference, vol. 1, no. 3, p. 6, [14] U. Karn, "An Intuitive Explanation of Convolutional Neural Networks," The Data Science Blog, [Online]. Available: [Accessed ]. [15] C. Ledig, et al. "PhotoRealistic Single Image SuperResolution Using a Generative Adversarial Network," arxiv, p. 19, 25 May [16] A.V. Nefian. Georgia Tech Face Database. Nov. 15, [Online]. Available: [Accessed: Nov. 5, 2017]. [17] Y.D. Wong. ChokePoint Dataset. [Online]. Available: arma.sourceforge.net/chokepoint/. [Accessed: Nov. 5, 2017].